Automated mail processing systems have been contemplated to process a high volume of mail at ever increasing speeds with the aim of reducing cost. Automated mail processing systems generally process mail by sorting incoming mail and match mailing outgoing mail. Mail sorting typically relies on the address or the code either printed on the outside face of the envelope or visible through a window provided in the envelope. Information printed on the outside the envelope or visible from the outside is typically read using an electronic reading device such as a relatively expensive optical character recognition (OCR) device, an image scanner, a bar code reader, a magnetic code reader, etc.
In automated match mailing processing systems, as described for example in U.S. Pat. No. 4,800,505 to Axelrod, an inserting apparatus or inserter is used with a carrier or insert having a machine readable code printed thereon corresponding to a match mailing identification that can be used to determine the items to be included in the outgoing mail. Similarly, U.S. Pat. No. 4,908,768 issued to Gelfer et al. uses a printer to imprint a machine readable code commonly known as a "dash code" onto a mail insert. The dash code typically represents the zip code of the addressee. A controller formats the information relating to each piece of mail obtained from the code and commands a printer to print out a manifest of the type and form that is acceptable to the mail service provider.
Alternatively, the documents themselves to be mailed can be marked with a machine code such as a dash code. The dash code information can then be automatically scanned for further processing. For example, the scanned information can be used to select and assemble the preselected documents into a single envelope for further processing, as described for example in U.S. Pat. No. 4,571,925 issued to Adams.
U.S. Pat. No. 4,796,196 to Durst, Jr. et al. further describes an automated outgoing and incoming mail processing system. In this system, a return stationery, such as a bill, and a return envelope with a code identifying the addressee printed thereon are generated or formatted for the outgoing mail. A conventional sorting module is used to read the preprinted code on the return envelopes of the incoming mail to automatically process them.
In this regard, U.S. Pat. Nos. 4,445,635 issued to Barr and 3,933,094 issued to Murphy et al. also describe a use of return envelopes having a machine readable code such as an optical character recognition, color coding or magnetic marking preprinted on the outside face of the envelope that can be read by a mail processing machine. Further, U.S. Pat. Nos. 5,267,754 issued to Kaule and 3,652,830 issued to Kessler describe postage stamps with a machine readable marking printed or coated thereon that can be attached to any envelope for automatic mail processing.
Automatic mail processing systems typically rely on some form of a machine code that is visible on the outside face of the envelope or visible therethrough. If the code is not visible or missing, then the systems cannot sort automatically and it is necessary to resort to the conventional time-consuming manual sorting and/or processing. It is desirable to automatically process mail even in situations where the codes are not visible or are otherwise missing. In this regard, U.S. Pat. No. 5,288,994 issued to Berson describes an image detecting apparatus and method for reading and/or verifying the contents in the sealed envelopes. The '994 patent detects the contents in the sealed envelopes by optically reading pre-encoded marks formed on the content, such as a return insert or carrier. A rather expensive image analysis system is required to reconstruct the image of the pre-encoded mark or to identify the pattern of the contents such as checks, business forms, payment stubs, etc. Moreover, to enable the analysis system to read through the sealed envelopes, the envelopes have to be transparent to various frequencies of light, especially near an infrared region.
It is common for mail order businesses and related industries to generate sales by mailing promotional packages, usually in very large volumes, to groups of predetermined customers or new prospects. For example, various mail order businesses offer sweepstakes prize awards on their promotional mailings to increase the recipients' attention to their product offerings. Specifically, these businesses invite customers to order products such as magazines and offer them the chance to enter a sweepstakes contest by mailing back the preaddressed courtesy return envelope provided to them. The customers that promptly return their order/entry document are entered into the sweepstakes contest regardless of whether any product is ordered.
These promotional mailings usually generate a tremendous volume of customer responses which must be promptly and accurately processed. Weekly volumes of return mail from promotional mailings can total in the millions during peak seasons and must be processed expeditiously to provide timely delivery of products and entry into the current sweepstakes awards. Some known current return mail processing consists of a combination of automated and manual procedures that are labor intensive. Automated bar code reading equipment is used to separate incoming customer response mail into various categories, basically sorting the returns generated by each different promotional mailing into separate groups. All return envelopes are then put through automated sorters, such as a Docutronix machine, which, in a single pass, top slit each envelope and separate envelopes containing contest entries into an order batch and a non-order batch. The Docutronix machine performs this separation by detecting the presence of magnetic ink placed on the order coupons. However, this machine is subject to an error factor of about 4%. Although all customer responses are subsequently processed for contest entry, the order batch must be separated for order processing.
While the two aforementioned steps enable a relatively quick separation of return envelopes into various categories, they only represent a small portion of the labor required to handle customer responses. Most notable are the ensuing processes of taking out the contents from the preslit envelopes and preparing their contents for data capture via either a high-speed OCR scanning device or manual key entry. Such process of taking out the content of each envelope and preparing for further processing is defined as "outsorting." The labor cost for this manually performed function alone is in the millions of dollars annually.
The direct mail industry and its related industries, are faced with ever increasing costs and the need to compress the time frames required to process huge volumes of customer responses. These costs include those related to machine-sorting return envelopes, manually outsorting their contents and then capturing the customer's preassigned ID number for entry into the contest and manually processing any magazine/product orders. Given the significant expense and time-consuming nature of the manual outsorting process described above, as well as the time and costs involved in the subsequent scanning and manual key entry processes that follow, there is a need for automating incoming mail processing. In particular, it is highly desirable to identify the customer associated with each return envelope and process their response mail without ever opening the envelope. This would provide a significant savings in labor and a competitive advantage. Furthermore, it is particularly desirable to automatically distinguish between orders and non-orders, without having to first open the return envelopes and to replace manual outsorting to reduce cost and increase speed.
Although some gains could be realized in the short term by using return envelopes with die-cut windows that would expose order, non-order and sweepstakes numbers, all of which can be scanned without first sorting or opening on current optical scanning optical devices, the savings would be minimal in the long run. Additionally, if the contestant does not properly enclose the insert, the order/non-order and contest numbers would not be detected, thus still requiring expensive manual outsorting.